Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    potential energy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dissipated
    0.66
     dissipation
    0.61
     dissip
    0.60
     dissipate
    0.59
     energet
    0.56
    energy
    0.55
     energy
    0.55
    熱
    0.54
     енер
    0.53
     heat
    0.52
    POSITIVE LOGITS
     Potential
    1.41
    Potential
    1.35
     potential
    1.27
     потенциа
    1.27
    potential
    1.19
     potencial
    1.13
     потен
    1.10
     potenc
    1.06
     potensi
    0.98
     potenciales
    0.97
    Act Density 0.017%

    No Known Activations